Every word of a speech triggers a midi note
Hello,
I have a question for the collective wisdom. I'd like to have every word of a speech to trigger one midi note that lasts until the next word. Pitch and vel. should be extracted from the analysis of the word . I used analyzer to get the pitch and the velocity (through the cooked output) but I seem to have lots of notes coming out all the time. I think there are lots of problems here (stop the notes from being retriggered. It is impossible to have one word one note for me, and analyses continues to analyze and send in midi notes...
Any idea what I do wrong?
Have you looked at the "attack" output of analyzer~?
The idea is that it sends a bang when a new note starts. But since it is expecting music, rather than speech, I suspect it won't be entirely what you need. You might be able to do better looking at the noisiness and amplitude of the signal. This is actually quite a hard problem.
here is a little amplitude thingy that use a lot to trigger events. maybe it will work for you...